410 CHAPTER 5 Another issue that complicates measuring the effectiveness of cross-sectional medication optimisation on long-term clinical outcomes, is that multimorbid older people with polypharmacy are in a rather dynamic health condition involving changes in (the severity of) medical conditions and pharmacotherapy over time [63]. These changes and subsequent actions by usual care interfere with clinical outcomes measuring drug-related harm. In Chapter 4.3, we found that the prior in-hospital medication review did not address medication errors identified at readmission because either these medication errors occurred after the medication review (~50%), no recommendation was given during the medication review (~25%) or the recommendation was not implemented (~25%). Thus, the relationship between reducing potentially inappropriate prescribing by a medication review (intervention) and the occurrence of a ‘DRA’ (outcome) over time is indirect and, therefore, difficult to measure. The type of medication optimisation recommendations and their implementation can highly affect treatment outcomes, given that each treatment has a unique benefit-risk balance. In addition, outcome measures in older people with polypharmacy and multimorbidity are highly affected by changes in medical conditions, pharmacotherapy and treatment goals over time, inadequately addressed by a single, cross-sectional in-hospital medication review. 3.2. How to measure the impact of medication review Prior to the initiation of the OPERAM trial, Beuscart et al. developed a core outcome set containing seven outcomes to be included in future trials of medication reviews in multimorbid older patients with polypharmacy: 1) drug-related hospital admissions; 2) drug overuse; 3) drug underuse; 4) potentially inappropriate medications; 5) clinically significant drug-drug interactions; 6) health-related quality of life and 7) pain relief [64]. Similarly, Rankin et al. developed a core outcome set including 16 outcomes to be considered in effectiveness studies aimed at improving the appropriateness of polypharmacy in older people in primary care. This core outcome set included endpoints in several domains: medication-related outcomes (e.g. adherence, medication appropriateness, medication errors, medication complexity), patient-related outcomes (e.g. cognitive functioning, quality of life, patient perception of treatment burden), patient’s knowledge about treatment and disease, healthcare utilisation and clinical (adverse) outcomes (e.g. severe drugrelated harm, mortality). These core outcome sets were thoroughly developed by Delphi consensus procedures and included the relevant outcomes that we attempted to improve via a medication review. However, given the highly individual medication review process, we learned from OPERAM and similar large clinical trials that it is challenging to measure the effect of a medication review on serious clinical adverse outcomes
RkJQdWJsaXNoZXIy MTk4NDMw